GET-Evidence

This is the only remaining step of the genome processing that uses MySQL queries. If we want to be able to send processing jobs out to virtual machines, it should just needs some links to files, not a local MySQL instance. Also, these MySQL queries are slow, too, changing this is likely to greatly speed up this processing step (currently around 30-40% of total). Another cases of removing MySQL queries sped up processing nearly 40-fold (23968b27)

Probably the thing to do is use "latest-flat.tsv", which should be regularly generated by a cron job from the GET-Evidence MySQL database.